inst/extdata/wx/DearWXpub/README.md

In kstawiski/OmicSelector: OmicSelector - a package for biomarker selection based on high-throughput experiments.

DearWXpub

wx keras version for everyone

Web tool : https://wx.deargendev.me/

We show the feature selection and Case/Normal classification result on TCGA dataset

A Keras implementation of Wx in preprint, : [Wx: a neural network-based feature selection algorithm for next-generation sequencing data Sungsoo Park, Bonggun Shin, Yoonjung Choi, Kilsoo Kang, and Keunsoo Kang] (https://www.biorxiv.org/content/biorxiv/early/2017/11/18/221911.full.pdf)

Differences with the paper: - We use the Keras as Neural Network running framework, original paper used naive tensorflow framework - Selected Features can be different as which backend learning framework used - Some TCGA samples updated, so we have more samples than paper.

Contacts - Your contributions to the repo are always welcome. Open an issue or contact me with E-mail [e-mail]

Usage

Step 1.

Experienment Environments

$ python 3.4
$ tensorflow gpu 1.4.0
$ keras 2.1.2

Install instructions

$ pip install tensorflow-gpu
$ pip install keras

Step 2. Clone this repository to local.

$ git clone https://github.com/deargen/DearWXpub.git
$ cd DearWXpub

Step 3. Download the TCGA RNA-seq data

1. Download rna-seq via TCGA-Assembler tool

2. we downloaded tool (Module_A.R) via http://www.compgenome.org/TCGA-Assembler/index.php

3. Run tcga_download.R

4. you may have to install releative packages

 - Shall Terminal
 $ sudo apt-get install libssl-dev
   * deb: libssl-dev (Debian, Ubuntu, etc)
   * rpm: openssl-devel (Fedora, CentOS, RHEL)
   * csw: libssl_dev (Solaris)
   * brew: openssl@1.1 (Mac OSX)

 $ sudo apt-get install libcurl4-openssl-dev
   * deb: libcurl4-openssl-dev (Debian, Ubuntu, etc)
   * rpm: libcurl-devel (Fedora, CentOS, RHEL)
   * csw: libcurl_dev (Solaris)

- R-studio
 > install.packages('curl')
 > install.packages('httr')
 > install.packages('stringr')
 > install.packages('rjson')

• you can see the 'TCGA_DATAS' folder in current DearWXpub path.

• Data Status

| 　 | TOTAL | Tumor | Normal | Tumor Ratio(%) | |------|-------|-------|--------|----------------| | TYPE | 6226 | 5609 | 617 | 90.09 | | BLCA | 427 | 408 | 19 | 95.55 | | BRCA | 1214 | 1101 | 113 | 90.69 | | COAD | 327 | 286 | 41 | 87.46 | | HNSC | 566 | 522 | 44 | 92.23 | | KICH | 90 | 65 | 25 | 72.22 | | KIRC | 606 | 534 | 72 | 88.12 | | KIRP | 323 | 291 | 32 | 90.09 | | LIHC | 424 | 374 | 50 | 88.21 | | LUAD | 576 | 517 | 59 | 89.76 | | LUSC | 553 | 502 | 51 | 90.78 | | PRAD | 549 | 497 | 52 | 90.53 | | THCA | 571 | 512 | 59 | 89.67 |

Step 4. Do the Feature selection and Get Classification Accuracy

$ python src/wx_tcga.py

It will generate preprocessed TCGA data set. And Select features, Get the scores.

Results

Selected Gene Markers

Keras Wx 14 ['EEF1A1','FN1','GAPDH','SFTPC','AHNAK','KLK3','UMOD','CTSB','COL1A1','GPX3','GNAS','ATP1A1','SFTPB','ACTB']

Peng 14 ['KIF4A','NUSAP1','HJURP','NEK2','FANCI','DTL','UHRF1','FEN1','IQGAP3','KIF20A','TRIM59','CENPL','C16ORF59','UBE2C']

with WX ranking ['KIF4A'(12970),'NUSAP1'(15886),'HJURP'(11479),'NEK2'(11939),'FANCI'(17123),'DTL'(12600),'UHRF1'(11825), 'FEN1'(17497),'IQGAP3'(14173),'KIF20A'(13057),'TRIM59'(11113),'CENPL'(10344),'C16ORF59'(9463),'UBE2C'(9039)]

edgeR 14 ['LCN1','UMOD','AQP2','PATE4','SLC12A1','OTOP2','ACTN3','KRT36','ATP2A1','PRH2','AGER','PYGM','PRR4','ESRRB']

with WX ranking ['LCN1'(3847),'UMOD'(7),'AQP2'(106),'PATE4'(5878),'SLC12A1'(165),'OTOP2'(7785),'ACTN3'(4939),'KRT36'(9441), 'ATP2A1'(14193),'PRH2'(2155),'AGER'(934),'PYGM'(13688),'PRR4'(2151),'ESRRB'(9630)]

Case Classifiation Accuracy

TCGA data( Downloaded at Dec. 26th. 2017 ), Half of data for feature selection / Half of data for validation

Accuracy of 14 Biomarker

|       |         | Wx 14 |       | Peng 14 |       | EdgeR 14 |       | |:-----:|---------|--------|-------|---------|-------|----------|--------| | TYPE | SAMPLES | Hit | Acc(%)| Hit | Acc(%)| Hit | Acc(%) | | TOTAL | 3119 | 3017 | 96.72 | 2961 | 94.93 | 2957 | 94.81 | | BLCA | 214 | 205 | 95.79 | 208 | 97.20 | 203 | 94.86 | | BRCA | 608 | 597 | 98.19 | 586 | 96.38 | 558 | 91.78 | | COAD | 164 | 155 | 94.51 | 143 | 87.20 | 162 | 98.78 | | HNSC | 283 | 275 | 97.17 | 261 | 92.23 | 267 | 94.35 | | KICH | 46 | 44 | 95.65 | 44 | 95.65 | 46 | 100.00 | | KIRC | 303 | 302 | 99.67 | 293 | 96.70 | 301 | 99.34 | | KIRP | 162 | 161 | 99.38 | 158 | 97.53 | 161 | 99.38 | | LIHC | 212 | 192 | 90.57 | 201 | 94.81 | 186 | 87.74 | | LUAD | 289 | 283 | 97.92 | 282 | 97.58 | 286 | 98.96 | | LUSC | 277 | 272 | 98.19 | 268 | 96.75 | 275 | 99.28 | | PRAD | 275 | 257 | 93.45 | 260 | 94.55 | 254 | 92.36 | | THCA | 286 | 274 | 95.80 | 257 | 89.86 | 258 | 90.21 |

WX 14 AUC curve (LUSC, LUAD, BRCA)

Accuracy of 7 Biomarker

Keras wx 7

'EEF1A1', 'FN1', 'GAPDH', 'SFTPC', 'AHNAK', 'KLK3', 'UMOD'

Martinez 7

BLCA : 'SMAD2', 'RUNX2', 'ABTB1', 'ST5', 'CEBPB', 'SETDB1', 'CEBPG'

BRCA : 'JAK2', 'NFKBIA', 'TBP', 'RXRA', 'VAV1', 'HES5', 'NFKBIB'

HNSC : 'DUSP16', 'KRT8', 'RAF1', 'MED1', 'PPARG', 'YWHAB', 'FABP1'

KIRC : 'AR', 'HGS', 'RUNX1', 'BCL3', 'BRCA1', 'STAT2', 'ITGA8'

LUAD : 'DOK1', 'FUT4', 'INSR', 'ITGB2', 'SHC1', 'PTPRC', 'KHDRBS1'

LUSC : 'BRCA1', 'ETS2', 'HIF1A', 'JUN', 'LMO4', 'PIAS3', 'RBBP7'

| 　 | 　 | WX 7 | 　 | Martinez 7 | 　 | |-------|------|------|---------|------------|--------| | TYPE | SAMPLES | Hit | Acc(%)| Hit | Acc(%)| | TOTAL | 3119 | 2986 | 95.74 | 　 | 　 | | BLCA | 214 | 205 | 95.79 | 206 | 96.26 | | BRCA | 608 | 591 | 97.20 | 556 | 91.45 | | COAD | 164 | 152 | 92.68 | 　 | 　 | | HNSC | 283 | 269 | 95.05 | 268 | 94.70 | | KICH | 46 | 45 | 97.83 | 　 | 　 | | KIRC | 303 | 299 | 98.68 | 273 | 90.10 | | KIRP | 162 | 162 | 100.00 | 　 | 　 | | LIHC | 212 | 187 | 88.21 | 　 | 　 | | LUAD | 289 | 283 | 97.92 | 260 | 89.97 | | LUSC | 277 | 271 | 97.83 | 257 | 92.78 | | PRAD | 275 | 249 | 90.55 | 　 | 　 | | THCA | 286 | 273 | 95.45 | 　 | 　 |

Wx example from actual data file

$ python src/wx_examples.py

Wx example for k > 2

$ python src/wx_gse_multi_class.py

Additional Validation

GSE72056

Case type : melanoma

Sample size
malignant : 1257
normal : 3256
total : 4513

Dataset splits(trn,dev,tst) = (3611, 902)
We perfomred the 5 fold cross validation

Accuracy Result (test set)
WX14 : 90.71%
Peng14 : 70.22%

GSE40419

Case type : luad(lung)

Sample size
malignant : 87
normal : 77
total : 164

Dataset splits(trn,dev,tst) = (132,32)
We perfomred the 5 fold cross validation

Accuracy Result(test set)
WX14 : 80.00%
Peng14 : 56.87%

GSE103322

Case type : Head and Neck(sinle cell)

Sample size
malignant : 2215
normal : 3687
total : 5902

Dataset splits(trn+dev,tst) = (4722,1180)
We perfomred the 5 fold cross validation

Accuracy Result(test set, 1180)
WX14 : 81.10%
Peng14 : 68.28%

kstawiski/OmicSelector documentation built on April 10, 2024, 11:11 p.m.